Chest pain is one of the most frequent reasons for calls to emergency medical services (EMS) (Colbeck, 2016). More than 57 000 Australians experience a myocardial infarction every year, resulting in 21 deaths each day (Heart Research Institute (HRI), 2025). Despite this, approximately only 10% of patients presenting with chest pain obtain a final diagnosis of acute coronary syndrome (ACS) (Napoli, 2017); of these, many present without electrocardiogram (ECG) changes (Khan et al, 2017). The spectrum of potential pathologies makes coming to a definitive diagnosis challenging in the prehospital setting.
Prehospital clinicians are trained to use predictive tools in the assessment of ACS (Nehme et al, 2013). Both the HEART (Backus, 2024) and TIMI (Antman, 2024) scores are examples of predictive tools that have been developed through cardiovascular research and are most beneficial when used within a hospital setting. While these predictive tools have demonstrated effectiveness in calculating the risk of a future major adverse cardiac event (MACE), research into whether they could be amended to become diagnostic tools in the prehospital setting is limited.
The use of cardiac biomarkers, specifically troponin assessment, adds a further assessment tool into the acute management pathways for these patients with ACS (Stengaard et al, 2017; Alghamdi and Body, 2018). A rapid initial assessment in conjunction with other predictive tools can triage patients based on their MACE risk level, allowing for more efficient patient movement within hospital. Including other predictive measures alongside prehospital troponin testing may assist paramedic services and clinicians in executing an appropriate risk analysis.
Therefore, the purpose of this research is to identify the gaps within the current scientific literature surrounding the predictive measures associated with MACE. It is anticipated that gaps in the literature related to the use of predictive tools will reveal prospects for further research.
This review also looks to evaluate the capacity of predictive tools to forecast MACE and improve the ability to diagnose ACS in the prehospital setting when used in conjunction with cardiac biomarkers.
Methods
A scoping review was performed using the Joanna Briggs Institute (JBI) population/concept/context (PCC) framework to identify the main concepts of the primary review question while focusing on the target population and context (Peters et al, 2020a).
The inclusion criteria were formulated using the PCC framework. Filtering of articles that were produced using the search terms followed the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) extension for scoping reviews (Tricco et al, 2018). PRISMA provides the essential reporting items for scoping review production and gives an overview of the source-selection process.
The study protocol was registered in the Open Science Framework registries network (Bampton et al, 2023).
Population
The population in the review included both male and female patients presenting with chest pain that was presumed to be cardiac in nature.
Patients anticipated to be experiencing ACS or non-ST-elevation myocardial infarction (NSTEMI) are the major focus for the review.
This review excludes those in which ST-segment elevation was identified on the ECG, as this finding is in most cases definitive in diagnosis of an acute event and is outside the conceptual investigations of this review. Articles relating to chest pain that was suspected to be pleuritic or traumatic in nature were also excluded.
Concept
The research concept involves the use of assessment tools used to create a predictive diagnosis of ACS in the absence of ST-segment elevation. This includes physical assessments and subsequent scoring, as well as assessment using devices, specifically point-of-care (POC) blood analysis.
Assessment tools used to predict short- and long-term mortality and MACE were the primary focus, where a MACE incorporates potential requirement for invasive cardiothoracic intervention, as well as mortality. Coronary artery bypass grafting (CABG) or percutaneous coronary intervention (PCI)/coronary angioplasty are the suggested invasive cardiothoracic interventions.
Context
Results strictly from prehospital settings make up the context, including paramedicine and its primary derivatives such as EMS and ambulance care.
Since hospitals are equipped with high-sensitivity troponin analysis and access to echocardiogram and angiogram equipment, article results from hospitals were used only to determine the diagnostic accuracy of prehospital assessment processes, and to acquire data surrounding mortality and MACE.
The review is not limited by geographical location but excludes studies published in languages other than English.
Types of sources
This scoping review considered both experimental and quasi-experimental study designs, including randomised controlled trials, non-randomised controlled trials and before-and-after studies. In addition, analytical observational studies, including prospective and retrospective cohort studies, case-control studies and analytical cross-sectional studies were considered for review and subsequent inclusion.
Qualitative studies were considered if the focus was on qualitative data including, but not limited to, designs such as phenomenology, ethnography and qualitative description.
In addition, systematic reviews that met the inclusion criteria were also considered, provided the research question was relevant by means of the PCC.
This review excluded descriptive observational study designs including case series, individual case reports, descriptive cross-sectional studies and service-audit reports. Grey literature was also excluded on the basis that more rigorous peer-reviewed data would be used.
Search strategy and criteria
Peer-reviewed articles published from January 2000 to August 2022 were obtained from searches of the two electronic databases, CINAHL Plus (with full text) and MEDLINE (Ovid). All identified articles were exported to the reference management software EndNote X9.
A PRISMA flow diagram was used to filter and identify the included studies following the eligibility criteria. The following were removed:
Articles that discussed both NSTEMI and STEMI were included, yet only data that involved NSTEMI were made available for discussion in line with the focus question. Potentially relevant sources were retrieved in full text and characterised through the JBI PCC framework (Peters et al, 2020b) to determine suitability.
A record of all included and excluded sources was kept ensuring appropriate register in the PRISMA extension to provide an overview of the screening process.
The results of the search were used in a scoping review so a broad search strategy was adopted while allowing reproducibility, reliability and transparency of the current state of literature. The search terms are listed in Table 1, and the PCC terms are summarised in Table 2.
Search terms applied to databases CINAHL Plus and MEDLINE (Ovid)
| ((MH “Acute Coronary Syndrome”) OR TI (NSTEMI OR “Non ST Segment Elevation Myocardial Infarction” OR ACS OR “Acute Coronary Syndrome”)) |
| AND |
| (MH (“Emergency Medical Technicians” OR “Prehospital Care) OR paramedic* OR ambulance OR EMT OR “emergency medical technician*” OR prehospital OR pre-hospital OR “out of hospital”) |
Summary of population/concept/context terms
| Population | Concept | Context |
|---|---|---|
| NSTEMI | Diagnosis | Paramedic |
| Non-ST segment elevation myocardial infarction | Mortality | Ambulance |
| ACS | Adverse cardiac events | Emergency medical technician |
| Acute coronary syndrome | MACE | Prehospital |
| Risk | Out-of-hospital |
Data organisation and synthesis
Data charting was developed by a sole reviewer. Variables that corresponded with the research aims and PCC terms were identified and included in a characteristics table (Table 3).
Summary of articles reviewed
| Authors | Country | Type of Study | Population | Concept/predictive tool use | Context | Findings/major outcomes |
|---|---|---|---|---|---|---|
| Price (2008) | UK | Quantitative | Not specified | Quality and performance of cardiac markers for rapid triage | Paramedics and hospital | Evidence suggests that POC troponin assessment using more sensitive assays are effective for ruling out AMI in accelerated triage protocols. There is less evidence suggesting that rule-in diagnosis has clinical or economic impact |
| Nehme et al (2013) | Australia | Quantitative: systematic review |
n=unknown |
Short-term clinical prediction models | EMS and hospital | Models provide diagnostic accuracy for short-term outcomes but variable population data is a methodological weakness |
| Zègre-Hemsey et al (2015) | US | Quantitative: retrospective analysis |
n=735 |
Prehospital ECG abnormalities versus normal prehospital ECG and subsequent outcomes | Paramedics | 9.3% of patients using emergency services had a completely normal ECG on assessment. These patients had a significantly lower incidence of adverse hospital outcomes |
| Colbeck (2016) | Australia | Scoping review | Age: ≥25 years | HEART, ANTCP, RSVP3 | Paramedics | Positive troponin levels have the single highest likelihood ratio in diagnosis of ACS. Most symptoms have little to no discriminatory value in the diagnosis of ACS |
| Stengaard et al (2017) | Denmark | Quantitative: observational, prospective study | n=962 Age: mean 66 years (95% CI (59.8–71.9)) | Evaluation of potential early triage based on hs-cTnT and copeptin blood samples in the prehospital setting | Paramedics | The combination of copeptin and hs-cTnT could allow AMI to be safely ruled out in 45% of patients |
| van Dongen et al (2020a) | Netherlands | Quantitative: prospective observational study |
n=700 |
HEART score, POC troponin and occurrence of MACE | Paramedics | The HEART score, which includes troponin levels, is accurate in identifying patients with low and intermediate to high risk of MACE |
| van Dongen et al (2020b) | Netherlands | Quantitative: prospective observational cohort study |
n=689 |
HEART score and POC troponin versus high-sensitive troponin T | Prehospital | The HEART score and POC troponin testing produced similar accuracy in predicting MACE compared to hs-cTnT |
| Al-Zaiti et al (2021) | US | Quantitative |
n=750 |
HEART, TIMI, GRACE, FRISC and PURSUIT clinical risk scores | EMS | Risk scores perform better in identifying chest pain owing to ACS than in predicting MACE. The HEART score is the most effective predictive tool |
| Stopyra et al (2021) | UK | Quantitative: prospective observational cohort study |
n=506 |
Prehospital troponin and HEART | EMS | EMS blood collection decreases the time to laboratory result availability. Earlier diagnosis of patients with elevated troponin can facilitate more rapid cardiology consultation, treatment and admission |
| Van Dongen et al (2021) | US | Quantitative |
n=699 |
Prehospital versus hospital ability to classify risk in patients with non-ST elevation ACS | EMS and hospital | In 74% of patients, hospital and prehospital risk classification using the HEART score is similar. Paramedics are more likely to overestimate the history and ECG components of the HEART score |
ACS: acute coronary syndrome; AMI: acute myocardial infarction; ECG: electrocardiogram; EMS: emergency medical services; hs-cTnT: high-sensitivity cardiac troponin test; MACE: major adverse cardiovascular event; POC: point of care
Data on article characteristics were collated. Characteristics that were collated included authors, year of publication, country/affiliation, journal, type of study, population, concept/predictive tool used, context and a summary of the findings and/or major outcomes.
Data synthesis was carried out to identify findings and major outcomes from each journal and summarise the correlations. This was completed using a basic qualitative content analysis (Pollock et al, 2023) to ensure that the review of qualitative data was kept descriptive in nature. This synthesis format is recommended by JBI scoping review guidance (Peters et al, 2020b).
Results
The initial literature search yielded a total of 234 results. After removal of 36 duplicate articles, as well as manual removal of 11 articles that weren't automatically excluded from the initial search because of their date of publication, 187 articles remained for initial screening. Because of university database licensing limitations, 33 articles were ‘abstract only’, for which the full text was not available. Ultimately, 154 articles were screened using the JBI PCC framework (Peters et al, 2020b), resulting in a final total of 10 articles for analysis. A breakdown of the PRISMA extension and filtering of the results is shown in Figure 1.
PRISMA extension (Tricco et al, 2018)
After analysis of the results obtained from PRISMA, summaries of each article are presented in Table 3. Two key themes were identified from the 10 articles: prehospital biomarker assessment and clinical risk scores.
From the qualitative content analysis, recurring findings that influenced MACE were highlighted. These included predictive factors, clinical risk scores, delay to definitive care and biomarker analysis.
As both clinical risk scores and biomarker analysis can influence MACE and have the potential to be more influential in the prehospital setting than predictive factors regarding delay to definitive care, these were selected for discussion. Both excluded themes are bound by uncontrollable variables to the treating clinician; predictive factors involve patient medical history and presenting symptomology, while geographical constraints, resourcing and staffing may influence delay to definitive care.
Risk-score models assist prehospital and in-hospital clinicians in their approach to ACS assessment and making a likely diagnosis, closing the gap between prehospital and hospital diagnosis discrepancies (Nehme et al, 2013; Al-Zaiti et al, 2019; van Dongen et al, 2020a; 2021).
It was noted that biomarker analysis in conjunction with risk scores can be used effectively in the prehospital setting if staff are adequately trained and a universal standard is adhered to (van Dongen et al, 2021). However, a significant lack of research was identified on the effectiveness of clinical risk scores and cardiac biomarkers when used in this way in the prehospital setting.
Discussion
The purpose of this research was to review the literature on predictive-assessment tools in the diagnosis of ACS and their potential to predict MACE in acute care medicine.
Through this review, gaps within the scientific literature were determined. This scoping review identified clinical risk scores and prehospital biomarker analysis as the key themes for discussion.
Clinical risk scores
Despite the effectiveness of risk-score models, prehospital diagnoses of ACS may differ or be less accurate than those made in hospital (Schewe et al, 2019). Furthermore, a risk-score matrix is used for provisional risk analysis. Clinical judgement is used to recognise signs, symptoms and history, and stratify the risk of MACE by recognising a high-risk patient through the score steps. As a result of this, clinical risk scores could be influenced by the clinical judgement of the user.
Comparison of clinical risk scores
Nehme et al (2013) acknowledged that a significant proportion of patients with suspected ACS would benefit from earlier triage using clinical prediction models and subsequent management at PCI-capable facilities.
Al-Zaiti et al (2019) noted that many of the risk scores and prediction models used in a clinical setting for patients with suspected ACS have not been validated for prehospital use. Therefore, comparisons of the HEART, TIMI, GRACE, FRISC and PURSUIT scores for chest pain owing to ACS was performed. Predictions of 30-day MACE were also completed for patients arriving through EMS. It was determined that clinical risk scores are more effective in ACS diagnosis than in prognosis and predictions of MACE (Colbeck, 2016; Poldervaart et al, 2017; Al-Zaiti et al, 2019).
In this review, HEART was shown to be the most effective clinical risk score in diagnosing or excluding the possibility of ACS (Al-Zaiti et al, 2019; van Dongen et al, 2021). Within a hospital setting, HEART also has a high non-predictive value (NPV) (>97%), where the NPV identifies the likelihood that an individual with a negative HEART score is later confirmed to not be experiencing ACS.
Therefore, the HEART score is sensitive in lower-risk patient subgroups, suggesting its capacity to identify low-risk patients who are less likely to experience MACE. With some modifications to the factors contributing to calculating a HEART score, this is considered a useful tool in prehospital ACS assessment and treatment pathways.
Clinical risk scores in prehospital settings
A significant proportion of patients with ACS will benefit from immediate triage and management at specialist, PCI-capable centres (Nehme et al, 2013). There, a case can be made that integrating a systems-based approach in managing patients with ACS by determining their risk of MACE would be appropriate.
Prehospital clinicians typically follow risk-mitigation strategies to enhance patient care. Further research is needed to determine whether, with the expansion of risk-scores, some patients with chest pain could be safely left at home with a referral to their primary care provider.
Several methods of risk mitigation have been validated for suspected ACS (Al-Zaiti et al, 2019). In current practice, paramedics have the opportunity for preliminary risk-stratification by classifying patients into subcategories of risk. These risk matrices have also been enhanced by the inclusion of prehospital biomarker analysis, specifically cardiac troponin T (cTnT) analysis, which assists the clinician in making more accurate diagnosis of suspected ACS (van Dongen et al, 2020b).
Prehospital cardiac biomarkers
Only recently has research evaluated the potential influence of prehospital clinicians taking and assessing blood samples on patient care (Alghamdi et al, 2019; Stopyra et al, 2021). Earlier diagnosis of ACS may be possible from prehospital biomarker analysis (Stengaard et al, 2013; 2017; Stopyra et al, 2020).
The development of accelerated triage processes and the literature surrounding early intervention for these patients indicates that the inclusion of a cardiac biomarker assessment pathway may be feasible in the prehospital setting.
Efficacy of prehospital troponin analysis
Price (2008) deduced that cTnT measurement during transport by ambulance could assist with the rapid triage of patients with chest pain through a rule-out process. Owing to the availability of POC devices, and cTnT demonstrating a higher specificity for myocardial ischaemia compared with myoglobin analysis, cTnT is the most feasible marker for prehospital assessment (Colbeck, 2016).
In instances where cTnT is not so unequivocally elevated, Price (2008) suggested that better analytical sensitivity is required to identify earlier stages of the development of ACS in the prehospital setting. In relation to this statement, studies have compared the diagnostic ability of hs-cTnT (high-sensitive cTnT) to that of cTnT. van Dongen et al (2021) compared the use of hs-cTnT to standard POC cTnT used in conjunction with the HEART score; they were found to have similar diagnostic capabilities.
Current guidelines recommend serial cTnT assessment when the onset of ACS symptoms is within 6 hours of presentation, owing to troponin elevations typically showing the greatest rise during the 2-to 12-hour period following myocardial insult (Babuin and Jaffe, 2005). This is supported by Stengaard et al (2013), who highlighted that POC cTnT assessment used in isolation significantly improved accuracy when symptom duration is >2 hours (Stengaard et al, 2013; van Dongen et al, 2020b). This is a challenging factor for prehospital clinicians as they are more commonly going to have patient contact at the acute phase of symptom onset. Unless the patient is located in a remote area with extended transport times, prehospital clinicians are unlikely to have patient contact for long periods of time. Further research is needed to identify if prehospital risk stratification is possible only if the patient's symptoms began >6 hours before contact.
From an operational perspective, given the time frames required to perform serial cTnT analysis, it may be feasible for prehospital clinicians to initiate blood draws for patients with suspected ACS, with the intention of carrying the analysis out in the prehospital setting. This would be preferable to handing the specimens to hospital staff for testing on arrival. Patients brought in by prehospital clinicians may benefit from an earlier diagnosis of ACS should their cTnT have a value suggestive of ACS as outlined by the likelihood ratios (LRs) identified by Colbeck (2016). This means that their treatment pathway can be streamlined or they can be safely discharged. Further research would be required in this field to determine operational and economic feasibility.
High-sensitivity troponin in conjunction with copeptin analysis
The level of hs-cTnT measured in prehospital blood samples from patients with suspected ACS has significant diagnostic and prognostic value. (Stengaard et al, 2017). Greater diagnostic accuracy can be achieved when copeptin analysis is included as well (Stengaard et al, 2017). With evidence to suggest measuring copeptin has improved prognostic classification, the combination of copeptin and hs-cTnT measurement could be used to safely rule out 45% of suspected ACS patients upon hospital arrival (Stengaard et al, 2017). However, copeptin is not used in the prehospital setting.
Gu et al (2011) noted that early concentration peaks of copeptin may allow ACS to be identified very soon after symptom onset. This was consistent with the findings of Stengaard et al (2017), who found increased circulating copeptin levels in patients with ACS experiencing the shortest symptom duration.
With copeptin and hs-cTnT analysis not yet being possible in a POC format, their influence on patient care is limited to analysis upon arrival to hospital. Therefore, they are unlikely to effect change to prehospital care in the near future. Nonetheless, the NPV demonstratinsg the accuracy of copeptin analysis for patients with hyperacute symptoms suggests that it could be a candidate for future research on its effectiveness within prehospital settings. With POC technology constantly evolving, a future where selected cardiac biomarkers are assessed at scene may develop.
Limitations
Limitations in the present research include that the PRISMA approach used for this scoping review was performed in July 2022, so literature published after then is not included. The PRISMA diagram shows that 33 reports and articles were abstract-only and not available in full text. Historical perspectives may also be invalidated by more modern research as the publication year for the PRISMA search was set from January 2000.
During analysis, copeptin was shown to be a new line of research and was not considered in the initial search terms. The search term ‘biomarker’ produced noteworthy results that included copeptin. However, results may have been more extensive had copeptin been used as a search term.
Future directions
Despite not yet having prehospital validation, the HEART score demonstrates potential for prehospital use in assisting the clinician in the diagnosis and prediction of MACE. HEART has been determined to be the most effective clinical risk score to date in diagnosing or excluding the possibility of ACS. Owing to its notable NPV, has the potential to accurately identify patients who fall into the low-risk subset of ACS.
Troponin was highlighted as a significant improvement to the initial HEAR score and illustrates the usefulness of biomarkers when used in conjunction with the other factors in calculating risk scores (Uyan, 2023). It would be valuable to identify the effectiveness of copeptin in replacing troponin in risk scores, or perhaps working in conjunction with POC troponin in the field, should POC copeptin ever become available with modern technological advances. As copeptin appears to show more diagnostic value than POC troponin in more acute phases of ACS, this may be suited to prehospital use.
Conclusion
Clinical risk associated with diagnosis, mortality and MACE in ACS remains a focus in prehospital clinical practice to improve ACS management. Patients presenting with symptoms consistent with ACS in the prehospital arena are usually treated within a structured treatment pathway from the time of ambulance contact up to and including discharge from hospital. Symptomology alone is insufficient and not specific enough for a rule-in rule-out approach to ACS in the prehospital setting, which is not new information to clinicians.
There is an opportunity, however, to improve the approach that prehospital clinicians take to patients presenting with suspected ACS. This could be achieved by redeveloping assessment processes and questioning in ways that are more aligned with likelihood ratios, and integrating the findings from clinical risk scores and biomarker assessment.
The main literature gaps concerned the use of the biomarker copeptin in conjunction with other clinical risk strategies, as well as the use of predictive tools in the prehospital setting. While the use of predictive tools and clinical risk matrices are well documented within clinical settings for ACS (Colbeck, 2016; Al-Zaiti et al, 2019; van Dongen et al, 2020a; 2021), literature surrounding the prehospital use of risk scores is almost non-existent. This extends to the use of copeptin in conjunction with predictive tools, similar to the use of cTnT in the HEART score. This may have potential for future qualitative research.
Patients with a completely normal ECG upon initial contact with EMS have an almost negligible level of MACE (Turnipseed et al, 2010). Therefore, the normal ECG may also highlight the need for further research into how prehospital ACS is managed.